Lithium for Mild Cognitive Impairment: A Pilot Trial Shows Feasibility, Not Yet Proof
Pilot clinical trials serve a specific and limited purpose. They are designed to answer whether a treatment is safe enough and feasible enough to justify testing in a larger, more expensive study - not to deliver the final verdict on whether that treatment works. The distinction matters enormously when interpreting results, because a null finding in a pilot trial means something very different from a null finding in a fully powered phase III study.
A pilot randomized clinical trial of low-dose lithium in adults with mild cognitive impairment, published in JAMA Neurology, illustrates this precisely. The study, led by Dr. Ariel G. Gildengers at the University of Pittsburgh, set out to establish three things: that the trial design was feasible, that low-dose lithium was safe and tolerable in this population, and that the data could generate reliable effect size estimates for planning a future study. On all three counts, the trial succeeded. On the primary clinical outcomes, none reached the prespecified significance threshold.
What the Trial Was Designed to Find
The scientific rationale for testing lithium in mild cognitive impairment comes from observational evidence suggesting that long-term lithium use in older adults with bipolar disorder is associated with better brain integrity compared with those who never took the drug. That correlation raised the question of whether lithium's apparent neuroprotective properties could benefit people at risk for Alzheimer's disease - a population distinct from those with bipolar disorder.
The trial enrolled adults aged 60 and older who had been clinically diagnosed with mild cognitive impairment, a condition involving noticeable memory or cognitive changes that do not yet meet criteria for dementia. Participants were randomized to low-dose lithium or placebo and followed for two years with cognitive testing, brain imaging, and biomarker assessments.
The Results and Their Limits
None of the trial's coprimary outcomes - the specific cognitive and neuroimaging measures that were prespecified as primary endpoints - showed a statistically significant difference between the lithium and placebo groups at the threshold set before the trial began.
That null result, however, must be interpreted with an important caveat that the authors acknowledge directly: the trial was not powered to detect clinically meaningful differences. Pilot trials typically enroll far fewer participants than a definitive trial would require. They generate preliminary data - effect sizes and variance estimates - that inform the sample size calculations for the next study. The data from this trial now provide exactly that: numbers that investigators can plug into power calculations to design a properly sized follow-up trial.
The trial did confirm that low-dose lithium is safe in older adults with mild cognitive impairment when administered under careful clinical monitoring. That safety confirmation is a necessary prerequisite for proceeding to a larger trial and is itself a meaningful outcome.
What Comes Next
The research team has identified a key design change for the next study: enrolling participants based on amyloid beta status rather than clinical symptoms alone. When this trial was designed, blood-based tests for amyloid pathology were not yet clinically available. As a result, not all participants had the Alzheimer's-related biology that lithium's proposed mechanism would target. A future trial that uses blood biomarkers to select amyloid-positive participants would be testing lithium in the specific population most likely to show an effect, increasing the odds of detecting a real signal if one exists.